Defining Prognostic Features in Triple-Negative Breast Cancer

Abstract & Commentary

By William B. Ershler, MD, Editor, INOVA Fairfax Hospital Cancer Center, Fairfax, VA; Director, Institute for Advanced Studies in Aging, Washington, DC.

Synopsis: Breast cancers that are "triple-negative" (ie, ER, PR, HER2-negative) are considered high risk, yet there remains variability, with some demonstrating less malignant features than others. In an effort to further define this group, UK investigators performed additional laboratory assessments and correlated these with outcomes in a series of patients from their institution. In addition to established clinical parameters (tumor size, nodal status, etc.), of the additional tests, androgen receptor and cytokeratin basal phenotype were shown to offer the most useful prognostic value.

Source: Rakha EA, et al. Prognostic markers in triple-negative breast cancer. Cancer. 2007;109:25-32.

Breast cancers that are devoid of estrogen and progesterone receptors are more likely to be poorly differentiated, of higher histological grade, associated with a higher recurrence rate and decreased overall survival.1, 2 Nonetheless, there are certainly examples of hormone receptor beast cancers for which these aggressive features are not observed.3, 4 The absence of the HER-2 receptor is considered favorable because HER-2 positive tumors have a higher growth fraction and are associated with a higher recurrence rate and decreased overall survival.5 Yet, HER2 negative cancers are also very heterogeneous with regard to clinical course.

Rakha and colleagues from Nottingham, UK report an evaluation of a well-characterized series of breast cancer patients with particular attention to those with the "triple-negative" phenotype (ie, estrogen receptor-negative, progesterone receptor-negative, and HER2-negative). Their goal was to determine how best to stratify subsets and characterize proliferative potential based upon available microarray and immunohistochemical panels. Of the 1944 cases of invasive breast cancers observed at the Nottingham Tenovus Primary Breast Carcinoma Series between 1986 and 1998, 16.3% were of the triple-negative phenotype. The median OS survival for the whole series was 73 months and time of event-free survival was 66 months. The triple-negative phenotype was associated with larger size, higher grade, more frequent distant metastasis and an overall poorer prognosis. Additional studies performed on these samples included an assessment of expression of androgen receptor, epidermal growth factor receptor (EGFR), P-cadherin, E-cadherin, basal cytokeratin (CK5/6, CK14) and p53. Within the group of triple-negative tumors, nodal status, tumor size, and the presence of androgen receptor provided the greatest prognostic value. Additionally, associations were found with loss of expression of E-cadherin and gain in expression of basal cytokeratins (basal phenotype), P-cadherin, p53 and EGFR. Of the 282 triple negative cases, 178 were LN negative and 104 were LN positive. In the LN positive group, size and androgen receptor expression had significant prognostic value. In the LN negative group, however, only the basal phenotype (cytokeratin expression), which was associated with a more negative outcome, demonstrated prognostic value. The authors concluded that for those patients found to have "triple negative" breast cancer, further assessment to determine androgen receptor expression (favorable) and cytokeratin "basal phenotype" should be undertaken to refine prognosis and develop treatment strategy.


Breast cancer is typically classified by tumor size, histological features, nodal status, hormone receptor status and the presence or absence of distant metastases. Under most circumstances this is sufficient information to formulate rational treatment strategies. Yet, there remains variability in outcome, and this presumably in some way is related to the variability in malignant properties of the cancer cell. This, coupled with the availability of new technology allowing a more detailed classification of gene expression, opens the possibility for developing an even more precise classification system; one that would more accurately predict tumor aggressiveness and response to therapy.6, 7 The current report is a step in that direction. Within the 'triple-negative' breast cancer phenotype there is variability with regard to clinical outcome. Much of this can be predicted by clinical parameters (nodal status, tumor size, etc.), but it appears that for those that express androgen receptor, clinical course might be expected to be somewhat more favorable. In contrast, expression of the basal cytokeratin phenotype would be considered unfavorable, particularly within the lymph-node negative group, a finding consistent with other studies.8, 9

Hopefully, these findings and others yet to come will not only provide prognostic information, but direction with regard to specific, targeted therapies.


1. Kinne DW, Butler JA, Kimmel M, Flehinger BJ, Menendez-Botet C, Schwartz M. Estrogen receptor protein of breast cancer in patients with positive nodes. High recurrence rates in the postmenopausal estrogen receptor-negative group. Arch Surg. 1987;122(11):1303-1306.

2. Pichon MF, Broet P, Magdelenat H, et al. Prognostic value of steroid receptors after long-term follow-up of 2257 operable breast cancers. Br J Cancer. 1996;73(12):1545-1551.

3. Jensen ML, Kiaer H, Andersen J, Jensen V, Melsen F. Prognostic comparison of three classifications for medullary carcinomas of the breast. Histopathology. 1997;30(6):523-532.

4. Trendell-Smith NJ, Peston D, Shousha S. Adenoid cystic carcinoma of the breast: a tumour commonly devoid of oestrogen receptors and related proteins. Histopathology. 1999;35(3):241-248.

5. Hartmann LC, Ingle JN, Wold LE, et al. Prognostic value of c-erbB2 overexpression in axillary lymph node positive breast cancer. Results from a randomized adjuvant treatment protocol. Cancer. 1994;74(11):2956-2963.

6. Mattie MD, Benz CC, Bowers J, et al. Optimized high-throughput microRNA expression profiling provides novel biomarker assessment of clinical prostate and breast cancer biopsies. Mol Cancer. 2006;5:24.

7. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406(6797):747-752.

8. Malzahn K, Mitze M, Thoenes M, Moll R. Biological and prognostic significance of stratified epithelial cytokeratins in infiltrating ductal breast carcinomas. Virchows Arch. 1998;433(2):119-129.

9. Potemski P, Kusinska R, Watala C, Pluciennik E, Bednarek AK, Kordek R. Prognostic relevance of basal cytokeratin expression in operable breast cancer. Oncology. 2005;69(6):478-485.